Trajectories of protein intake and 28-day mortality in critically ill patients: a secondary analysis of a cluster-randomized controlled trial

Abstract

Background & aims The optimal protein intake approach during the early phase of critical illness remains controversial. This study aimed to evaluate the association between different trajectories of protein intake and 28-day mortality in a cohort of critically ill patients. Methods The NEED trial is a multicenter, cluster-randomized controlled trial assessing the impact of an actively implemented evidence-based nutrition guideline on mortality in critically ill patients. This secondary analysis included the patients who stayed in ICU for at least seven days. Group-based trajectory modeling was applied to identify subgroups with similar protein intake trajectories in this cohort. Cox proportional hazards models were used to analyze the impact of different trajectories on 28-day mortality. Results Overall, 2191 patients were included for analysis. A distinct triple-group trajectory of protein intake was identified, with 919 patients categorized into the low-level protein intake group, 1146 the medium-level group, and 126 the high-level group. The mean daily protein intake from the low-to high-level protein intake group during the first week of enrollment were 0.38 ± 0.14, 0.8 ± 0.18, and 1.68 ± 0.39 g/kg/d, respectively. Compared with the medium-level protein intake group, the low-level or high-level protein intake group was associated with significantly increased 28-day mortality (hazard ratio [HR] = 1.348, 95% confidence interval [CI]: 1.067-1.704; HR = 2.291, 95% CI: 1.533-3.423, respectively). After controlling for potential confounders, the adjusted HRs were 1.365 (95% CI: 1.032-1.807) for the low-level group and 1.921 (95% CI: 1.274-2.896) for the high-level group. However, when taking energy intake into account, low-level protein intake was no longer related to mortality. In contrast, the detrimental effects of high-level protein intake remain tenable (Adjusted HR = 2.324, 95% CI: 1.524-3.543, P < 0.001). Conclusions Low-level or high-level protein intake in the early phase of critical illness was associated with increased 28-day mortality than medium-level protein intake. However, when adjusted for energy intake, low-level protein intake in the early phase was no longer associated with increased 28-day mortality.